Abstract
Due to the under-sparsity or over-sparsity, the widely used regularization methods, such as ridge regression and sparse representation, lead to poor hallucination performance in the presence of noise. In addition, the regularized penalty function fails to consider the locality constraint within the observed image and training images, thus reducing the accuracy and stability of optimal solution. This paper proposes a locally weighted sparse regularization method by incorporating distance-inducing weights into the penalty function. This method accounts for heteroskedasticity of representation coefficients and can be theoretically justified from Bayesian inference perspective. Further, in terms of the reduced sparseness of noisy images, a moderately sparse regularization method with a mixture of ℓ1 and ℓ2 norms is introduced to deal with noise robust face hallucination. Various experimental results on public face database validate the effectiveness of proposed method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Freeman, W., Pasztor, E., Carmichael, O.: Learning low-level vision. Int. J. Comput. Vis. 40(1), 25–47 (2000)
Baker, S., Kanade, T.: Limits on super-resolution and how to break them. IEEE Trans. PAMI 24(9), 1167–1183 (2002)
Chang, H., Yeung, D.Y., Xiong, Y.M.: Super-resolution through neighbor embedding. In: CVPR, pp. 275–282 (2004)
Bevilacqua, M., Roumy, A., Guillemot, C., Alberi, M.L.: Low-complexity single-image super-resolution based on nonnegative neighbor embedding. In: British Machine Vision Conference (BMVC), Surrey, pp. 1–10 (2012)
Ma, X., Zhang, J., Qi, C.: Hallucinating face by position-patch. Pattern Recogn. 43(6), 3178–3194 (2010)
Park, J.-S, Lee, S.-W: An example-based face hallucination method for single-frame, low-resolution facial images. IEEE Trans. Image Process. 17(10), 1806–1816 (2008)
Suo, F., Hu, F., Zhu, G.: Robust super-resolution reconstruction based on adaptive regularization. In: WCSP, pp. 1–4 (2011)
Liu, C., Shum, H., Freeman, W.: Face hallucination: theory and practice. Int. J. Comput. Vis. 7(1), 15–134 (2007)
Li, Y., Lin, X.: An improved two-step approach to hallucinating faces. In: Proceedings of IEEE Conference on Image and Graphics, pp. 298–301 (2004)
Tang, Y., Yan, P., Yuan, Y., Li, X.: Single-image super-resolution via local learning. Int. J. Mach. Learn. Cyber. 2, 15–23 (2011)
Jiang, J., Hu, R., Han, Z: Position-patch based face hallucination via locality-constrained representation. In ICME, pp. 212–217 (2012)
Fan, J., Li, R.: Variable selection via nonconcave penalized likelihood and its oracle properties. J. Am. Stat. Assoc. 96(456), 1348–1360 (2001)
Yang, J., Tang, H., Ma, Y., Huang, T.: Image super-resolution via sparse representation. IEEE Trans. Image Process. 19(11), 2861–2873 (2010)
Zeyde, R., Elad, M., Protter, M.: On single image scale-up using sparse-representations. In: Curves and Surfaces, pp. 24–30 (2010)
Jung, C., Jiao, L., Liu, B., Gong, M.: Position-patch based face hallucination using convex optimization. IEEE Signal Process. Lett. 18(6), 367–370 (2011)
Jia, Z., Wang, H., Xiong, Z.: Fast face hallucination with sparse representation for video surveillance. In: ACPR, pp. 179–183 (2011)
Zhang, J., Zhao, C., Xiong, R., Ma, S., Zhao, D.: Image super-resolution via dual-dictionary learning and sparse representation. In: ISCAS, pp. 1688–1691 (2012)
Meinshausen, N., Bühlmann, P.: High-dimensional graphs and variable selection with the Lasso. Ann. Statist. 34(3), 1436–1462 (2006)
Li, Q., Lin, N.: The Bayesian elastic net. Bayesian. Analysis 5(1), 151–170 (2010)
Zou, H., Hastie, T.: Regularization and variables election via the elastic net. J. R. Stat. Soc. B 67(2), 301–320 (2005)
Candès, E.J., Wakin, M.B., Boyd, S.: Enhancing sparsity by reweighted ℓ1 minimization. J. Fourier Anal. Appl. 14(5), 877–905 (2008)
Friedlander, M.P., Mansour, H., Saab, R., Yılmaz, O.: Recovering compressively sampled signals using partial support information. IEEE Trans. Inf. Theory 58(2), 1122–1134 (2012)
FEI Face Database. http://fei.edu.br/~cet/facedatabase.html
Acknowledgments
This work was supported by the National Natural Science Foundation of China (61231015, 61172173, 61170023, 61303114, 61271256, U1404618), the Fundamental Research Funds for the Central Universities (2042014kf0286, 2042014kf0212, 2042014kf0025, 2042014kf0250), China Postdoctoral Science Foundation (2014M562058), and Natural Science Fund of Hubei Province (2015CFB406).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Wang, Z., Xiao, J., Lu, T., Shao, Z., Hu, R. (2015). Tuning Sparsity for Face Hallucination Representation. In: Ho, YS., Sang, J., Ro, Y., Kim, J., Wu, F. (eds) Advances in Multimedia Information Processing -- PCM 2015. PCM 2015. Lecture Notes in Computer Science(), vol 9314. Springer, Cham. https://doi.org/10.1007/978-3-319-24075-6_29
Download citation
DOI: https://doi.org/10.1007/978-3-319-24075-6_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-24074-9
Online ISBN: 978-3-319-24075-6
eBook Packages: Computer ScienceComputer Science (R0)